Acquired immunodeficiency syndrome (AIDS) dementia complex (ADC) has been characterized as a subcortical manifested by cognitive, motor, and behavioral impairment that eventually affects at least two-thirds of patients with AIDS. The cause of ADC is unknown. Histopathological examinations of mild to moderate cases of ADC reveal no obvious involvement of neurons, astrocytes, or oligodendrocytes. Our initial studies supporting this application demonstrated that quantitative P-31 NMR spectroscopy detects biochemical changes at a cellular level in the brain that are not manifest in conventional MRI. High-energy phosphate concentrations were depleted about 30% in patients with ADC when compared to normal controls, without a corresponding loss of brain tissue due to atrophy. This project will test the hypotheses that: 1. The severity of neurological dysfunction in AIDS Dementia Complex increases with the decrease in brain High-energy phosphate metabolite concentrations and 2. Decreased High-energy phosphate metabolite concentrations will predict the onset of AIDS Dementia not manifested by traditional proton NMR Imaging (MRI) or neuropsychological evaluations. We will test these hypotheses by studying 60 HIV positive patients, 30 symptomatic and 30 asymptomatic of ARC, by quantitative P-31 NMR spectroscopy, H-1 NMR imaging (MRI), and neuropsychological evaluations according to the 1990 neuropsychological battery recommendations of the NIMH. Each patient will be examined by all three non-invasive modalities (NMR spectroscopy, imaging and neuropsychological tests) every 24-26 weeks for two years, for a total of five examinations. The results from the HIV patients will be compared to identical examinations, (NMR spectroscopy, imaging, and neuropsychological evaluations) of 30 healthy control subjects at risk for HIV infection, performed every 24-26 weeks for two years. The neuropsychological evaluations and the NMR imaging will act as clinical standards against which the spectroscopy findings will be evaluated to determine the efficacy and reliability of P-31 NMR spectroscopy as a diagnostic tool. We will methodologically assess the progression of AIDS dementias and compare the temporal biochemical, neuroanatomical, and neuropsychological changes to normal control subjects. We will determine the relationship between high-energy phosphate metabolite concentrations and the severity of neurological dysfunction associated with AIDS dementia, and by statistical analyses determine whether the changes in high-energy phosphates predict one or more changes in neuropsychological function.